Conversion of Biomass to Charcoal and th

The Holocene 17,4 (2007) pp. 539–542

Conversion of biomass to charcoal
and the carbon mass balance from
a slash-and-burn experiment in
a temperate deciduous forest
Eileen Eckmeier,1* Manfred Rösch,2 Otto Ehrmann,3
Michael W.I. Schmidt,1 Wolfram Schier4 and Renate Gerlach5
(1University of Zurich, Department of Geography, Winterthurerstrasse 190, 8057 Zurich,
Switzerland; 2Landesamt für Denkmalpflege im Regierungspräsidium Stuttgart, Fischersteig 9,
78343 Gaienhofen-Hemmenhofen, Germany; 3Münster 12, 97933 Creglingen, Germany; 4Freie
Universität Berlin, Institut für Prähistorische Archäologie, Altensteinstrasse 15, 14195 Berlin,
Germany; 5Landschaftsverband Rheinland, Rheinisches Amt für Bodendenkmalpflege, Endenicher
Strasse 133, 53115 Bonn, Germany)
Received 21 July 2006; revised manuscript accepted 19 December 2006

Abstract: Anthropogenic burning, including slash-and-burn, was deliberately used in (pre)historic Central
Europe. Biomass burning has affected the global carbon cycle since, presumably, the early Holocene. The
understanding of processes and rates of charcoal formation in temperate deciduous forests is limited, as is
the extent of prehistoric human impact on the environment. We took advantage of an experimental burning to simulate Neolithic slash-and-burn, and we quantified the biomass fuel and charcoal produced, determined the resulting distribution of the charcoal size fractions and calculated the carbon mass balance.
Two-thirds of the charcoal particles (6.71 t/ha) were larger than 2000 ␮m and the spatial distribution of

charcoal was highly variable (15–90% per m2). The conversion rate of the biomass fuel to charcoal mass
was 4.8%, or 8.1% for the conversion of biomass carbon to charcoal carbon, and 58.4 t C/ha was lost during the fire, presumably as a component of aerosols or gases.
Key words: Slash-and-burn experiment, temperate deciduous forest, charcoal, carbon mass balance,
biomass burning.

Introduction
The timing of the onset of anthropogenic influence on the
global climate system is still under discussion and is possibly
connected to anthropogenic fires (Carcaillet et al., 2002;
Ruddiman, 2003). Palaeobotanical records suggest that anthropogenic burning was common in the past and may have been
used as a tool for hunting, herding and farming.
Anthropogenic fire occurred with high spatial and temporal
variation (Pyne, 1994). In temperate central Europe, burning
for landscape management and agriculture lasted from the prehistoric Mesolithic until the modern nineteenth century. Most
*Author for correspondence (e-mail: eckmeier@geo.unizh.ch)

© 2007 SAGE Publications

evidence for (pre)historic fire-clearance husbandry comes from
alpine regions and the pre-alpine midlands (Clark et al., 1989;

Rösch, 1993; Haas, 1996; Erny-Rodmann et al., 1997;
Carcaillet, 1998; Tinner et al., 2005), Scandinavia (Iversen,
1941; Kalis and Meurers-Balke, 1998; Hörnberg et al., 2006)
and Great Britain (Mason, 2000; Blackford et al., 2006).
The fossil charcoal records, however, do not necessarily
reflect former burning and land-use systems. Experimental field
studies on slash-and-burn could improve our understanding of
the processes involved in anthropogenic burning in temperate
deciduous forests. They should include budgets for biomass
fuel, charcoal and carbon to infer taphonomical implications
such as conversion rates from biomass to charcoal, charcoal
size distribution and spatial heterogeneity. But, to the best of

10.1177/0959683607077041

Downloaded from http://hol.sagepub.com at Universitaet Zuerich on July 8, 2007
© 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

540

The Holocene 17 (2007)

our knowledge, neither biomass nor charcoal budgets were
taken into account in previous slash-and-burn experiments.
We took advantage of a burning experiment in southwest
Germany (Forchtenberg) in a deciduous forest. The objectives
of our study were: (i) to quantify the biomass fuel and charcoal
produced during a typical slash-and-burn event in a temperate
deciduous forest, (ii) to determine the distribution of the produced charcoal in size fractions, and (iii) to calculate the carbon mass balance.

Carbon and nitrogen concentrations
and mass balance

Materials and methods

Results and discussion

Site description

Before the fire, we estimated a woody aboveground biomass of

278 t/ha on the site, a common value for temperate deciduous
wood (180–600 t/ha; Schulze et al., 2002). We used 131 t/ha of
wood fuel (dry mass) to burn the plot. The moisture content of
the wood was 34 vol.%. Grass and herbaceous vegetation comprised a very small proportion of the total fuel mass (Table 1).
This is the first slash-and-burn experiment to record the
amount of fuel and the mass of charcoal produced in deciduous forest, and a comparison with data available from natural
wildfires is difficult. The amount of fire-consumed biomass
varies from ecosystem to ecosystem and fire to fire (Schimmel
and Granström, 1996). As examples, burning logging slash
(conifer or eucalyptus wood) consumed 140 t/ha, a fire in primary tropical forest 77–228 t/ha (Stocks and Kaufmann, 1997).
The amount of wood fuel used during our burning experiment
fits well with the above-mentioned amount of burnt logging
slash, and it is also in the range of primary tropical forest.
After the fire, the distribution of the remaining biomass on
the plot was heterogeneous. The coverage of the area with
charcoal after the fire varied between 15 and 90% per m2. The
material remaining on the plot after burning was dominated by
fragments larger than 2000 ␮m (6.71 ⫾ 0.66 t/ha, standard
error), compared with the smaller fractions (0.31 ⫾0.04 t/ha in
1000–2000 ␮m), as shown in Figure 1. The amount of charred

material reflects the amount of consumed fuel. Natural boreal
forest fires in Scandinavia produced 0.235 t/ha of charcoal
(Ohlson and Tryterud, 2000) and a high-intensity fire in Siberia
generated 0.735 t/ha of airborne charred particles (Clark et al.,
1998). A clearing fire in the Amazonian rainforest produced
4.3 t/ha charcoal (Fearnside et al., 2001). Charcoal yields
obtained in our burning experiment were larger probably
because the burning technique used here involved larger quantities of small-wood fuel and the burning was less intensive
than in the wildfires.
The conversion rate was calculated using the total dry biomass and the charcoal mass ⬎2000 ␮m. We found that 4.8% of
the biomass fuel was converted to charcoal, and this is similar
to that from fires in other ecosystems. Novakov et al. (1997)
estimated that generally less than 10% of biomass is converted
to charred carbon during wildfires. Wildfires in boreal forest
converted 2.0% (Clark et al., 1998) or 2.2% (Lynch et al., 2004)
of biomass to charcoal, while an intense crown fire converted
8.0% of fuel to visually identified charcoal (Tinker and Knight,
2000). Slash-and-burn in Amazonian tropical forests converted
1.3–2.9% of fuel carbon to charcoal carbon (Fearnside et al.,
2001). The conversion rates differ owing to different burning

and sampling conditions; however, the rate of 4.8% measured
in our experiment and the rates from Amazonian rainforest
and boreal forests are relatively close.
Both organic carbon and nitrogen concentrations increased
during the charring of biomass. The average carbon concentration in the wood samples was 463 ⫾1.5 g C/kg. Before slashing,
the carbon stock of the total woody biomass on the plot was

The experimental slash-and-burn was designed to mimic
Neolithic agricultural slash-and-burn in order to asses the
effects on vegetation, crop yields and soil properties (Rösch
et al., 2002). The site is located near Forchtenberg (SW
Germany; 49°16' N, 09°28' E) in a temperate deciduous forest.
Fagus, Carpinus and Acer dominate the forest composition and
the undergrowth species are characteristic for a woodruff-beech
forest (Galio-Fagetum). The trees are about 40 years old and the
area has been forested for at least two centuries. The site is
3.5 ha in area; it is slightly sloping and exposed to the south
(320 m a.s.l.). The mean annual temperature is 8.9°C and the
mean annual precipitation is 849 mm. The soil is a Haplic
Luvisol (WRB) with partly stagnic properties. Soil moisture

was 30 vol.% on the day of burning.

Burning technique
The trees were cut down within a plot of 30 m ⫻ 30 m (April
2004), the trunks and larger branches (diameter ⬎10 cm) were
removed from the site. The small wood pieces were left to dry
until an area 11 m ⫻ 8 m was burnt in October 2004. A pile of
burning wood was drawn over the ground with long hooks and
was continually fed with wood. This technique was used so
that the grass and herbaceous vegetation could burn completely and to distribute the charcoal and fertile ash as homogeneously as possible. Similar techniques are known from
historical slash-and-burn agriculture in central European
mountainous regions; they were described by Schmithenner
(1923) and were used in former slash-and-burn experiments
(Reynolds, 1977).

Determination of biomass fuel and produced
charcoal masses
We determined the amount of dry biomass by (i) weighing the
total amount of wood used for burning and measuring its
water content, and (ii) collecting the homogeneously distributed grass and herbaceous plants from four different 1 m2 plots

located outside of the burning field and weighing it after drying in an oven at 40°C for three days.
After the burning, the distribution of charcoal particles on
the soil surface was visually estimated. We sampled 20 replicates of charcoal and unburnt aboveground biomass using a
frame (0.2 m ⫻ 0.2 m) and a vacuum cleaner. Each sample was
collected in a separate vacuum cleaner bag. Samples were wetsieved and passed through the mesh sizes 2000 ␮m, 1000 ␮m,
500 ␮m, 250 ␮m and 125 ␮m. The material ⬍125 ␮m, including
mineral soil material, was collected and dried. After drying the
samples at 40°C for 48 h, they were weighed.
Charcoal particles were separated from the unburnt material for
all 20 samples of the fraction ⬎2000 ␮m and for seven randomly
selected samples of the fraction 1000–2000 ␮m. The charred and
uncharred fractions were subsequently weighed. We defined
charcoal as black particles completely charred on the surface.

Total carbon and nitrogen concentrations were determined for
20 charcoal samples ⬎2000 ␮m, 20 samples ⬍125 ␮m and for
uncharred biomass samples (beech wood and grass material)
by dry combustion via an elemental analyser (Elementar
VarioEL). The measured carbon and nitrogen concentrations
were used to calculate the carbon and nitrogen mass balance

for biomass fuel and resulting charcoal.

Downloaded from http://hol.sagepub.com at Universitaet Zuerich on July 8, 2007
© 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

Eileen Eckmeier et al.: A slash-and-burn experiment in a temperate deciduous forest

541

Table 1 Carbon and nitrogen concentrations (g/kg) and carbon and nitrogen stocks (t/ha) with standard errors in biomass fuel and in postfire above-ground charcoal and uncharred biomass samples
Material

Particle
size
(␮m)

n

Mass
(t/ha)

Total woody biomass
before slashing
Wood fuel
Grass fuel
Charcoal
Uncharred
Charcoal
Uncharred
Total
Total
Total

⬎2000
⬎2000
1000–2000
1000–2000
500–1000
250–500
125–250

4
20
20
7
7
20
20
20

278
131
9.24 ⫾ 0.37
6.71 ⫾ 0.66
1.87 ⫾ 0.16
0.31 ⫾ 0.04
0.10 ⫾ 0.02
0.66 ⫾ 0.04
0.63 ⫾ 0.04
0.81 ⫾ 0.08

Total

⬍125

20

0.35d

Carbon

n

2
2
20

20

Nitrogen

Conc.
(g/kg)

Stocks
(t/ha)

Conc.
(g/kg)

Stocks
(t/ha)

–
463 ⫾ 2
437 ⫾ 0
775 ⫾ 1
–
–
–
–
–
–

129a
61a
4.04b
5.20c
0.86a
0.24c
0.05a
–
–
–

–

0.5 ⫾ 0.3
23.5 ⫾ 0.5
6.7 ⫾ 0.2
–
–
–
–
–
–

0.15a
0.07a
0.22b
0.04c
0.00a
0.00c
0.00a
–
–
–

144 ⫾ 0

–

5.0 ⫾0.2

–

The stocks were calculated using the nitrogen and carbon concentrations and the masses of samples. All size fractions contained charcoal;
the fractions ⬎2000 and 1000–2000 ␮ m were visually sorted into charred and uncharred particles.
Averages with standard error.
a Calculated with wood C or N concentration.
b Calculated with grass C or N concentration.
c Calculated with charcoal C or N concentration.
d Mass of soil mineral matrix was calculated and substracted from total.
– Not determined.

129 t C/ha. About one-half of the wood material (containing
61 t C/ha) was burnt, together with herbaceous and grass vegetation (4 t C/ha). Nitrogen concentrations in the wood charcoal
(6.7 ⫾ 0.2 g N/kg) were larger than in the initial wood
(0.5 ⫾0.2 g N/kg), showing the preferential enrichment of nitrogen during charring (Almendros et al., 2003). After the fire,
5.2 t C/ha was left in charcoal particles ⬎2000 ␮m and 0.9 C
t/ha in the uncharred debris. Thus, 8.1% of the biomass fuel
carbon (wood fuel and grass) was converted to charcoal carbon
and 1.3% was left uncharred. We calculated a loss of carbon
through aerosols and gases during one fire of 58.4 t C/ha, or
90.6% of the biomass fuel carbon. The calculated mass of emitted CO2 during the fire was 206 t CO2/ha, using the emission

factor for CO2 from extratropical forest fires (1569 g CO2/kg
dry fuel burnt) published by Andreae and Merlet (2001).
Several limitations to our study should be considered: (i) fires
are highly variable. Even our attempts to burn homogeneously
led to a heterogeneous distribution of charcoal. The forest at
the research area is not a primary forest but a young secondary
forest, and the first Neolithic settlers may have encountered
denser forests holding more biomass. Nevertheless, we could
quantify the amount of biomass fuel and the associated charcoal yield for a fire in a typical deciduous forest. (ii) We assume
that in our experiment no large charcoal particles (⬎500 ␮m)
were carried away by heat exturbations or wind during the
burning (Patterson et al., 1987). (iii) The elemental composition of charcoal may differ and depends on the plant tissue
from which the charcoal particles are derived and on the charring procedure that occurred.

Conclusions

Figure 1 Distribution of litter layer biomass stocks (t/ha) after
slash-and-burn in the different size classes (in ␮m). The fractions
⬎2000 and 1000–2000 ␮m were visually sorted into charred and
uncharred particles

Experimental slash-and-burn in a temperate deciduous forest
converted 4.8% of the forest biomass to charcoal ⬎2000 ␮m, or
8.1% of the initial biomass fuel carbon to charcoal carbon.
During the fire, 58.4 t C/ha was lost as a component of aerosols
or gases.
After the burning, the spatial distribution of charcoal was
highly variable because the fire, although controlled, did not
burn homogeneously. Thus, for representative results, both in
the recent and even more in the fossil terrestrial record, a large
number of sample replicates is a prerequisite.
The burning produced mainly charcoal particles ⬎2000 ␮m
(6.71 t/ha), which may imply that macrocharcoal would dominate the sedimentary charcoal records. However, before inferring amounts of burnt biomass from the sedimentary charcoal
records, one should further investigate the fate of charcoal
after its formation, a field where little quantitative information
is available.

Downloaded from http://hol.sagepub.com at Universitaet Zuerich on July 8, 2007
© 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

542

The Holocene 17 (2007)

Acknowledgements
We thank the Forchtenberg team members and supporting
staff for their assistance in the field, notably Reiner Lubberich
for his technical support. We also thank Ivan Woodhatch and
Karen Hammes for improving the English and helpful comments, Christoph Hartkopf-Fröder for stimulating discussions
and the reviewers for improving the manuscript. Financial support from the University of Zurich, Switzerland, is gratefully
appreciated.

References
Almendros, G., Knicker, H. and González-Vila, F.J. 2003:
Rearrangement of carbon and nitrogen forms in peat after progressive thermal oxidation as determined by solid-state 13C- and
15N-NMR spectroscopy. Organic Geochemistry 34, 1559–68.
Andreae, M.O. and Merlet, P. 2001: Emission of trace gases and
aerosols from biomass burning. Global Biogeochemical Cycles 15,
955–66.
Blackford, J.J., Innes, J.B., Hatton, J.J. and Caseldine, C.J. 2006:
Mid-holocene environmental change at Black Ridge Brook,
Dartmoor, SW England: a new appraisal based on fungal spore
analysis. Review of Palaeobotany and Palynology 141, 189–201.
Carcaillet, C. 1998: A spatially precise study of Holocene fire history, climate and human impact within the Maurienne valley,
North French Alps. Journal of Ecology 86, 384–96.
Carcaillet, C., Almquist, H., Asnong, H., Bradshaw, R.H.W.,
Carrión, J.S., Gaillard, M.J., Gajewski, K., Haas, J.N., Haberle,
S.G., Hadorn, P., Müller, S.D., Richard, P.J.H., Richoz, I., Rösch,
M., Sánchez Goñi, M.F.S., von Stedingk, H., Stevenson, A.C.,
Talon, B., Tardy, C., Tinner, W., Tryterud, E., Wick, L. and Willis,
K.J. 2002: Holocene biomass burning and global dynamics of the
carbon cycle. Chemosphere 49, 845–63.
Clark, J.S., Merkt, J. and Müller, H. 1989: Post-glacial fire, vegetation, and human history on the northern Alpine Forelands,
south-western Germany. Journal of Ecology 77, 897–925.
Clark, J.S., Lynch, J., Stocks, B.J. and Goldammer, J.G. 1998:
Relationships between charcoal particles in air and sediments in
west-central Siberia. Holocene 8, 19–29.
Erny-Rodmann, C., Gross-Klee, E., Haas, J. N., Jacomet, S. and
Zoller, H. 1997: Früher ‘human impact’ und Ackerbau im Übergangsbereich Spätmesolithikum-Frühneolithikum im schweizerischen Mittelland. Jahrbuch der Schweizerischen Gesellschaft für
Ur- und Frühgeschichte 80, 27–56.
Fearnside, P.M., Graça, P. and Rodrigues, F.J.A. 2001: Burning of
Amazonian rainforests: burning efficiency and charcoal formation
in forest cleared for cattle pasture near Manaus, Brazil. Forest
Ecology and Management 146, 115–28.
Haas, J. N. 1996: Pollen and plant macrofossil evidence of vegetation change at Wallisellen-Langachermoos (Switzerland) during
the Mesolithic–Neolithic transition 8500 to 6500 years ago. Dissertationes Botanicae 267.
Hörnberg, G., Bohlin, E., Hellberg, E., Bergman, I., Zackrisson,
O., Olofsson, A., Wallin, J.E. and Passe, T. 2006: Effects of
Mesolithic hunter-gatherers on local vegetation in a non-uniform
glacio-isostatic land uplift area, northern Sweden. Vegetation
History and Archaeobotany 15, 13–26.

Iversen, J. 1941: Landnam i Danmarks Stenalder. En pollenanalytisk Undersøgelse over det første Landbrugs Indvirkning paa
Vegetationsudviklingen. (Land occupation in Denmark’s Stone
Age. A pollen-analytical study of the influence of farmer culture
on the vegetational development). Danmarks Geologiske Undersølgelse Raekke II, 66.
Kalis, A.J. and Meurers-Balke, J. 1998. Die ‘Landnam’-Modelle
von Iversen und Troels-Smith zur Neolithisierung des westlichen
Ostseegebietes – ein Versuch ihrer Aktualisierung. Prähistorische
Zeitschrift 73, 1–24.
Lynch, J.A., Clark, J.S. and Stocks, B.J. 2004: Charcoal production, dispersal, and deposition from the Fort Providence experimental fire: interpreting fire regimes from charcoal records in
boreal forests. Canadian Journal of Forest Research-Revue
Canadienne de Recherche Forestière 34, 1642–56.
Mason, S. 2000: Fire and Mesolithic subsistance – managing oaks
for acorns in northwest Europe? Palaeogeography, Palaeoclimatology, Palaeoecology 164, 139–50.
Novakov, T., Cachier, H., Clark, J.S., Gaudichet, A., Macko, S. and
Masclet, P. 1997: Characterization of particulate products of biomass combustion. In Clark, J.S., Cachier, H., Goldhammer, J.G.
and Stocks, B., editors, Sedimental records of biomass burning and
global change. Springer, 117–43.
Ohlson, M. and Tryterud, E. 2000: Interpretation of the charcoal
record in forest soils: forest fires and their production and deposition of macroscopic charcoal. Holocene 10, 519–25.
Patterson, W.A., III, Edwards, K.J. and Maguire, D.J. 1987:
Microscopic charcoal as a fossil indicator of fire. Quaternary Science
Reviews 6, 3–23.
Pyne, S.J. 1994: Maintaining focus: an introduction to anthropogenic fire. Chemosphere 29, 889–911.
Reynolds, P.J. 1977: Slash and burn experiment. Archaeological
Journal 134, 307–18.
Rösch, M. 1993: Prehistoric land use as recorded in a lake-shore
core at Lake Constance. Vegetation History and Archeobotany 2,
213–32.
Rösch, M., Ehrmann, O., Herrmann, L., Schulz, E., Bogenrieder,
A., Goldammer, J.P., Hall, M., Page, H. and Schier, W. 2002: An
experimental approach to Neolithic shifting cultivation. Vegetation
History and Archaeobotany 11, 143–54.
Ruddiman, W.F. 2003: The anthropogenic greenhouse era began
thousands of years ago. Climatic Change 61, 261–93.
Schimmel, J. and Granström, A. 1996: Fire severity and vegetation
response in the boreal Swedish forest. Ecology 77, 1436–50.
Schmithenner, H. 1923: Die Reutbergwirtschaft in Deutschland.
Geographische Zeitschrift 29, 115–27.
Schulze, E.-D., Beck, E. and Müller-Hohenstein, K. 2002:
Pflanzenökologie. Spektrum.
Stocks, B.J. and Kaufmann, J.B. 1997: Biomass consumption
and behavior of wildland fires in boreal, temperate, and tropical
ecosystems: parameters necessary to interpret historic fire
regimes and future fire scenarios. In Clark, J.S., Cachier, H.,
Goldhammer, J.G. and Stocks, B., editors, Sedimental records of
biomass burning and global change. NATO ASI Series, SpringerVerlag, 169–88.
Tinker, D.B. and Knight, D.H. 2000: Coarse woody debris following fire and logging in Wyoming lodgepole pine forests. Ecosystems
3, 472–83.
Tinner, W., Conedera, M., Ammann, B. and Lotter, A.F. 2005: Fire
ecology north and south of the Alps since the last ice age. The
Holocene 15, 1214–26.

Downloaded from http://hol.sagepub.com at Universitaet Zuerich on July 8, 2007
© 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.